Crossing The Brain's Barrier: Recent Advances And What Lies Ahead

By Neeraja Revi, Ph.D., postdoctoral scholar, Department of Neurological Surgery, Feinberg School of Medicine, Northwestern University

On March 25, 2026, the FDA approved a drug that the brain drug delivery field has been hoping for over the last three decades. Accelerated approval was granted to AVLAYAH (tividenofusp alfa-eknm), a biologic specifically engineered to cross the blood-brain barrier (BBB). This approval is significant because roughly 98% of small molecules and effectively all large molecules fail to enter the human brain at therapeutic concentrations. Before March, there were no other therapies approved in the U.S.; however, one treatment had been approved in Japan in 2021: JCR Pharma's pabinafusp alfa, an anti-transferrin-receptor enzyme fusion for Hunter syndrome. In the years leading up to these groundbreaking approvals, the receptor-mediated transcytosis brain shuttle playbook has grown and reached the clinic with a blast. Now, the pragmatic question for future applications is not whether to use a brain shuttle, it’s which shuttle, against which target, and for which patient.

The Extremely Selective Guestlist And The Ever So Amusing Location

The BBB is not a single membrane, rather, it’s a system. Endothelial cells in the cerebral microvasculature seal together at tight junctions formed by claudin-5, occludin, and zonula occludens proteins. They are wrapped in pericytes and astrocytic endfeet and overexpress efflux transporters, such as P-glycoprotein and BCRP, that actively pump foreign molecules into the blood. The result is an extremely selective barrier that permits fewer than 2% of small molecules through passive diffusion and effectively zero unmodified antibodies to reach therapeutic concentrations in the CNS.

Experts in the field know that this is not a new observation or idea. William Pardridge, MD, who, in the 1980s, founded the modern field of BBB drug delivery, has been defining it for 40 years. His pioneering research formed our understanding of the utilization of targeting molecules like insulin, transferrin proteins, or peptide conjugated liposomes as vehicles to direct delivery to the brain. The novel findings are showcased in the volume of clinical evidence that support the receptor-mediated transcytosis pathway. These therapeutic developments achieve drug delivery to the brain by mechanisms similar to the Grecian Trojan horse: they use native receptors, like the transferrin receptor (TfR1), to carry cargo across the endothelial cells into the brain. Until 2024, the evidence was preclinical or, at best, biomarker-based and as of this spring, its regulatory.

The Strategy That Worked

There are a few programs that deserve our focus. Denali Therapeutics built its platform on what it calls the TransportVehicle (TV), an engineered Fc fragment that binds TfR1 in a pH-dependent manner and dissociates once inside the brain. The platform is modular by design, meaning they can create several types of fusion transport vehicles to treat various diseases. Here are some of their recent developments:

• Enzyme transport vehicle (ETV)
  • This model is the basis for AVLAYAH for Hunter syndrome and for DNL126 for Sanfilippo syndrome type A.
• Oligonucleotide transport vehicle (OTV)
  • DNL628 reduces tau by targeting MAPT in Alzheimer’s disease and is now in Phase 1b clinical trials.
• Antibody transport vehicle (ATV)
  • DNL921 is in clinical preparation against amyloid-beta in Alzheimer's disease.
• Progranulin transport vehicle (PTV)
  • TAK-594/DNL593 was developed in a partnership with Takeda to treat GRN-related frontotemporal dementia.

The team’s success is evident by the AVLAYAH Phase 1/2 results, published in December. AVALAYH showed the ability to reduce cerebrospinal fluid (CSF) heparan sulfate by up to 90% by week 24, with the majority of the treated patients reaching CSF heparan sulfate levels within the range of unaffected individuals.

Roche took a different route in targeting the same receptor. Its Brainshuttle module is a Fab fragment that binds TfR1 and is genetically fused to a therapeutic antibody. Trontinemab, the Brainshuttle-enabled successor to the failed amyloid antibody, gantenerumab, has produced an interesting data set in its Alzheimer's studies. At the Alzheimer's Association International Conference (AAIC 2025), Roche reported that 91% of high dose trontinemab-treated patients achieved amyloid PET negativity by week 28, with amyloid-related imaging abnormalities with edema/effusion (ARIA-E) rates under 5%. For comparison, lecanemab and donanemab, both FDA-approved monoclonal antibodies, clear amyloid slower and at higher amyloid-related imaging abnormalities (ARIA) risk. The Phase 3 studies for trontinemab, TRONTIER 1 and 2, trials were initiated in late 2025, and Roche has also announced a preclinical Alzheimer's prevention program on top of these studies.

Harvard's Wyss Institute is the third honorable mention, with a very different structure than the first two. The Brain Targeting Program is a precompetitive consortium that uses a coalition of funding from Bristol-Myers Squibb, Eisai, Eli Lilly and Company, Lundbeck, GSK, ABL-Bio, Leal, Visterra, and other industry partners for developing antibody-fragment shuttles targeting TfR1, CD98hc, and multiple undisclosed transcytosis targets. As of late 2025, the program had completed six licensing deals (among them ABL-Bio, GSK, Leal, Lundbeck, and Visterra) and reports a 10- to 60-fold improvement in brain uptake over unmodified controls. The noteworthy point here is that it’s not just the platform that creates a successful therapeutic. It’s the pharma and academia collaborations that are willing to share infrastructure and knowledge to solve problems that none of them could have cracked alone, at least within the given timelines.

The Alternatives Are Real, But Narrower

While the molecular Trojan horse approach was galloping, a parallel strategy was also in the works. Magnetic resonance (MR)-guided focused ultrasound, combined with intravenous microbubbles, offers a mechanism of entry by opening the BBB transiently and reversibly at a sub-millimeter resolution. Lipsman and colleagues at Sunnybrook and Johns Hopkins University published the first clinical evidence in five Alzheimer's patients in 2018. Jin Woo Chang's group published their work in 2025, expanding the field by successfully completing repetitive bilateral frontal BBB openings. INSIGHTEC, which manufactures the ExAblate system, and Columbia, which has developed noninvasive neuronavigation-guided single-element transducers, are also running parallel Phase 1 and 2 programs utilizing similar technologies.

Focused ultrasound has real advantages: it is noninvasive, reversible, spatially precise, and agnostic to the therapeutic cargo. These advantages are coupled with the anticipation of near-term therapies being acute, region-specific delivery (chemotherapy to glioblastoma) or, more speculatively, repeated openings to accelerate amyloid clearance in patients already on antibody therapy. However, it has real limitations, too. Each session requires an MR scanner and a trained operator with the window for treatment after the scan lasting only hours. The financial burden on end users should also be considered, as it may not be comparable to the costs associated with intravenous biologics for chronic CNS diseases.

Further, intrathecal delivery, intranasal routes, and adeno-associated virus (AAV) vector strategies all continue to maintain their roles, particularly for diseases where the carrier itself is the cargo. Some examples of intrathecal delivery include Ionis Pharmaceuticals’ and Biogen's nusinersen (Spinraza) for SMA, an antisense oligonucleotide delivered into the cerebrospinal fluid, and Alnylam's mivelsiran (ALN-APP), an RNAi therapeutic against amyloid precursor protein now in trials for Alzheimer's disease and cerebral amyloid angiopathy.

On the AAV front, Novartis has developed onasemnogene abeparvovec (Zolgensma) for SMA, an AAV9 gene therapy that delivers a functional SMN1 gene in a single intravenous dose. Gene therapy development is also underway, with vectors utilizing the same doorways as the antibody shuttles. For example, Deverman's group at the Broad Institute engineered an AAV capsid, BI-hTFR1, that binds the human transferrin receptor and produced 40 to 50 times more gene expression in the brain than AAV9 in mice carrying the human receptor.

What This Means For The Next Generation Of Researchers

TfR1-enabled routes, whether fused to enzymes, antibodies, oligonucleotides, or peptides, are now a recognizable category with a regulatory precedent. Based on the recent approval, one would expect progress around enzyme replacement therapies for lysosomal storage disorders to accelerate first, followed by anti-amyloid and anti-tau inventions.

For academia, this leaves an open-ended research question of: What receptor-mediated advances can be made outside of the use of TfR1? In the current state, the receptor has been well studied, and, in fact, the space is a little bit too crowded. CD98hc, which the Wyss program is actively shuttling, has the potential to be the next wheelhorse. Another route already in the focus utilizes IGF1R. ABL Bio's Grabody-B, an IGF1R targeting shuttle (which doesn’t interfere with the normal IGF1 activation), carries its anti-alpha-synuclein bispecific antibody, ABL301, into the brain for treatment of Parkinson’s disease. This development has been licensed to Sanofi and more recently the shuttle Grabody-B was licensed to GSK for broader applications in the neurodegenerative disease spectrum.

Beyond that question, my personal interest would be in the identification of therapies that confer regional specificity (cortex versus hippocampus versus cerebellum or even narrow subdomains), cell-type specificity (neurons versus astrocytes versus microglia), and disease-state specificity (inflamed versus healthy endothelium). Current tools like BBB-on-chip systems and the single-cell endothelial atlases might aid in identifying these niche targets. And my bet is that the next decade of BBB targeted therapies will feature some level of the aforementioned specificity.

For young researchers, the major gap is engineering. While the biology of transcytosis is fairly understood, the field needs engineers who can modify factors like affinity, surface valency, size, toxicity, and pH dependency. These attributes are necessary for shuttle modules and allow for the release of cargo at the right time and place. One such notable engineering achievement is TXP1, a TfR1- targeted shuttle, developed by Ossianix. This shuttle is built from single-domain shark VNAR fragments and is significantly smaller than the classic IgG. TXP1 is engineered to avoid red blood cell toxicity seen with other transferrin-receptor shuttles.

Other than engineering, there is a need for computational biologists who can mine endothelial transcriptomes for the novel targets and clinical researchers who can design biomarker-driven Phase 1 trials in CNS diseases where motor and cognitive endpoints take years to read out.

A Look Forward

The next few years will provide answers to some of the field’s most pressing questions. First, will Roche's TRONTIER Phase 3 trials replicate the trontinemab Phase 2 findings in terms of the cognition biomarker results, the endpoint that ultimately matters? The ambitious goal is to prevent or reduce cognitive decline in at-risk Alzheimer’s disease individuals before it manifests its early symptoms. Second, can studies confirm that the advantages of MR-guided focused ultrasound outweigh the limitations of the therapy? Lastly, in the search of widely expressed targets, is the field missing out on potent targets capable of location specific uptake of carriers?

Outside of these pointed questions, the current approach of target-hunting may be overlooking the human variant factor. A recent Wyss preprint profiled 11 canonical shuttle targets in patients of normal health or those with neurodegenerative conditions: the transcytosis receptor TfR1, insulin receptors LDLR, LRP1, and LRP8, the solute carriers GLUT1 and CD98hc, IGF-1 receptor, the GPI-anchored ALPL and CA4, and the flippase TMEM30A. This study was conducted across 11 human cohorts and found consistent expression between brain regions, disease states, sexes, and ages, with 612 of the 631 comparisons showing no significant difference. What varied in every group and for every target, however, was abundance of receptors per individual. There was a two- to tenfold variation between the highest and lowest expressing patient.

This study has brought an important variable to the attention of the field. If brain shuttle delivery is dependent on the abundance of receptors, and those vary on an individual basis rather than a population one, it could be a potential explanation for the variable responses already reported for some of these therapies. This data makes a strong case for companion biomarker testing, similar to HER2 or EGFR testing in cancer patients, before choosing appropriate therapy. And hopefully these study results can inform the next wave of brain shuttle trials.

With AVLAYAH’s approval, nothing in the brain drug delivery field seems impossible now. The past 40 to 60 years have been spent failing at effectively crossing the brain's barrier, and suddenly we see a shift resulting in regulatory approvals. This culmination of work is worth paying attention to, worth being precise about the “why” of success, and worth drawing inspiration from those who succeeded.

Disclaimer: The opinions expressed are solely those of the author and do not reflect the official policy or position of Northwestern University. The author declares no competing interests relevant to this work.

About The Author

Neeraja Revi, Ph.D., is a postdoctoral scholar in the Department of Neurological Surgery at Northwestern University's Feinberg School of Medicine. Her focus lies in designing nanocarriers that cross the blood-brain and blood-retinal barriers. Her current work extends into modulating glial cells in mental health disorders and to stem-cell strategies for repairing corticospinal neurons after spinal cord injury. She earned her Ph.D. in biotechnology from IIT Hyderabad and holds two patents, and multiple publications. She has received the Keystone Future of Science (travel award) and an International Brain Barriers Society poster award for her work on designing a receptor targeted nanocarrier for treating mental health disorders.